ns-execution-gitlab-issue
Pass
Audited by Gen Agent Trust Hub on Oct 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from GitLab issue payloads which may contain malicious instructions.
- Ingestion points: Issue title, description, and comments are fetched using the
read_issueMCP tool and passed to execution agents in Phase 2 of the workflow. - Boundary markers: The instructions do not define or enforce strict boundary markers or delimiters when interpolating external issue content into prompts for delegated agents like
ns-autonomousorns-coder. - Capability inventory: The skill possesses significant capabilities including the ability to perform git operations (fetch, branch, push), create and manage merge requests via MCP tools, and update GitLab issue metadata such as status labels and time tracking statistics.
- Sanitization: There is no explicit description of sanitization, escaping, or validation of the remote issue content before it is processed by the AI agents.
Audit Metadata